Agent Skills

abicheck publishes one Agent Skill — a portable, triggerable package of native-compatibility expertise that a coding agent (Claude Code, GitHub Copilot, OpenAI Codex, Cursor, Gemini CLI) loads when a user asks a compatibility question in their own words.

The skills are named after the user's job, not after abicheck commands. Someone who has never heard of abicheck can ask "will this change break existing consumers?" and get a workflow that reaches for abicheck as its deterministic verification engine. See ADR-058 for why the portfolio is shaped this way.

Requires unreleased abicheck 0.6.x to run. You can install this skill (copying a directory works regardless of your abicheck version) against the latest published release, 0.6.0 — but it checks the installed abicheck version against its own declared range — >=0.6.0,<0.7.0 — and refuses to run rather than fail partway through, since several commands/options it drives postdate 0.5.0 (see "Prerequisites" below). The check runs abicheck --version — the executable actually on PATH when the skill runs, not a source checkout's pyproject.toml (the two can differ, e.g. a 0.6.x checkout with a separately installed 0.5.0 package). If that version isn't inside the 0.6.x range, the installed skill will decline to execute.

Portfolio status (2026-08-20): the portfolio was reset from four published skills down to one internal candidate. No skill has measured evidence yet that it improves agent behavior over a well-documented CLI alone, so publishing three additional, unvalidated skills alongside the one flagship was scaling packaging ahead of any validated product value. native-api-evolution, native-consumer-compatibility, and native-release-compatibility — previously published at prototype status — are no longer published; their source is recoverable from git history for whenever a second skill is built. check-abi-compatibility (renamed from review-native-library-change, itself renamed from native-binary-compatibility-review) is the sole surviving skill and is an internal candidate: working, reviewed content, but not yet validated and not for external publication or citation as validated in any user-facing claim, pending G37's still-unrun behavioral evaluation. See ADR-058's 2026-08-20 amendment for the full rationale (the same status is also kept in skills-src/CLAUDE.md, the skill's own source tree).

The catalogue

Skill	Status	The question it answers
check-abi-compatibility	Internal candidate	"Will this change break existing consumers?" — review a diff, branch, commit, or PR, ending in a verdict plus a root-cause explanation. Also handles "why did this suddenly report dozens of breaks?"

Installing them

skills-src/ is the one hand-authored source; the three publication trees below are generated build output, not committed to this repository (2026-08-21 ADR-058 amendment) — regenerate them with python scripts/gen_agent_skills.py (writes all three) or python scripts/install_dev_skill.py --target <name> (writes one or more by name: codex, claude, gemini, or all):

Tree	Read by
.agents/skills/	GitHub Copilot, OpenAI Codex, Cursor — the portable, cross-vendor convention
.claude/skills/	Claude Code, which does not scan .agents/skills
.gemini/skills/	Gemini CLI, which does not either

Each generated skill directory is fully self-contained: after generating it, copy .agents/skills/<skill-name>/ into your own project (or your personal skills directory) and every reference it needs comes with it. There are no symlinks and no cross-skill paths, so a single skill installs and works on its own.

Skills are executable content. Anthropic's own guidance applies to these as to any others: install only from sources you trust, and read what you install. The rules these skills hold themselves to — never manufacture a green result, never widen a suppression, never mutate a project silently — are in ADR-058's safety invariants, with the operational copy shipped inside every skill as references/shared/safety-invariants.md.

Prerequisites

The skills drive the abicheck CLI, so they need whatever the workflow they run needs — see CLI usage for installation and evidence and build-context flags for what deeper --depth levels require. Local CLI invocation is the only execution path a skill uses, so a skill works in any agent with shell access and needs no protocol server or other setup beyond installing the skill itself. (ADR-058 wrote this as "CLI normative, MCP an optional adapter"; abicheck has since removed its MCP server, so the CLI is simply the backend.)

Each skill declares, in its frontmatter, the abicheck version range it requires — the releases that actually provide the CLI surface it drives — and checks the installed version before doing anything else rather than degrading silently. The current portfolio targets 0.6.x: several commands and options these workflows depend on (aggregate, project plan, --view root-cause, --diagnostic-comparison, --contract) postdate the 0.5.0 release, so an older installation is refused up front instead of failing partway through a workflow.

Contributing

Edit skills-src/, never the generated trees. skills-src/CLAUDE.md is the contributor contract — the three-layer model, the shared-fragment rules, and the admission bar a second public skill would have to clear. The phased plan is G36.

python scripts/gen_agent_skills.py          # regenerate all three trees
python scripts/gen_agent_skills.py --check  # what CI gates on